In recent years, the problem of evaluating the trustworthiness of machine learning systems has become more urgent than ever. A directly related issue is that of assessing the fairness of their decisions. In this work, we adopt a primarily logical perspective on the topic, by trying to highlight the basic logical characteristics of the inferential setting in which a biased prediction occurs. To do so, we first identify and formalise four key desiderata for a logic capable of modelling the behaviour of a biased system, namely: skewness, dependency on data and model, non-monotonicity, and the existence of a minimal distinction between types of bias. On this basis, we define two metrics, one for group and one for individual fairness
Reasoning With Bias / C. Manganini, G. Primiero (CEUR WORKSHOP PROCEEDINGS). - In: Fairness and Bias in AI / [a cura di] R. Calegari, A. Aler Tubella, G. González Castañe, V. Dignum, M. Milano. - Kraków : Ceur, 2023 Oct 25. - pp. 1-16 (( Intervento presentato al 1. convegno Workshop on Fairness and Bias in AI tenutosi a Kraków nel 2023.
Reasoning With Bias
C. Manganini
Primo
;G. PrimieroUltimo
2023
Abstract
In recent years, the problem of evaluating the trustworthiness of machine learning systems has become more urgent than ever. A directly related issue is that of assessing the fairness of their decisions. In this work, we adopt a primarily logical perspective on the topic, by trying to highlight the basic logical characteristics of the inferential setting in which a biased prediction occurs. To do so, we first identify and formalise four key desiderata for a logic capable of modelling the behaviour of a biased system, namely: skewness, dependency on data and model, non-monotonicity, and the existence of a minimal distinction between types of bias. On this basis, we define two metrics, one for group and one for individual fairnessFile | Dimensione | Formato | |
---|---|---|---|
Reasoning With Bias.pdf
accesso aperto
Tipologia:
Publisher's version/PDF
Dimensione
375.26 kB
Formato
Adobe PDF
|
375.26 kB | Adobe PDF | Visualizza/Apri |
Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.